ARTICLE IN PRESS

Int. J. Production Economics 87 (2004) 251–266

Enterprise information systems project implementation:

A case study of ERP in Rolls-Royce
Yahaya Yusufa,*, A. Gunasekaranb, Mark S. Abthorpec
a
Business School, University of Hull, Hull HU6 7RX, UK
b
Department of Management, University of Massachusetts, 285 Old Westport Road, North Dartmouth, MA 02747-2300, USA
c
Department of Mechanical and Manufacturing Engineering, Nottingham Trent University, Burton Street, Nottingham NG1 4BU, UK

Abstract

Economic globalisation and internationalisation of operations are essential factors in integration of suppliers,
partners and customers within and across national borders, the objective being to achieve integrated supply chains. In
this effort, implementation of information technologies and systems such as enterprise resource planning (ERP)
facilitate the desired level of integration. There are cases of successful and unsuccessful implementations. The principal
reason for failure is often associated with poor management of the implementation process. This paper examines key
dimensions of implementation of ERP system within a large manufacturing organisation and identifies core issues to
confront in successful implementation of enterprise information system. A brief overview of the application of ERP
system is also presented and in particular, ERP software package known as SAP R/3, which was the ERP software
package selected by Rolls-Royce plc. The paper takes an in-depth look at the issues behind the process of ERP
implementation via a case study methodology. It focuses on business and technical as well as cultural issues at the heart
of the Rolls-Royce implementation. The case study also looks at the implementation time scales and assesses the
benefits from the project both tangible and intangible.
r 2003 Elsevier B.V. All rights reserved.

Keywords: ERP; Information systems; Implementation; Success factors; Rolls-Royce

1. Introduction quality in addition to greater flexibility and

The global nature of modern marketplace
requires active players to internationalise their
operations. In the past, companies were used to
competing based on one or two competitive
performance objectives such as price and quality.
However, present markets demand both price and
*Corresponding author. Tel.: +44-1486466933; fax: +44-
1482466216.
E-mail address: yahaya.yusuf@hull.ac.uk (Y. Yusuf).
responsiveness and thus today’s organisations
must compete based on all competitive objectives.
In order to achieve such simultaneity in perfor-
mance objectives, some organisations have decen-
tralised their operations by global outsourcing of
activities. This places enormous challenge on
companies to achieve a co-ordinated and inte-
grated supply chain. The emergence of various
information technologies such as the Internet,
electronic data interchange (EDI) and WWW
facilitate the attainment of an integrated supply

0925-5273/$ - see front matter r 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.ijpe.2003.10.004
 ARTICLE IN PRESS
252 Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
chain and in turn flexibility and responsiveness in
meeting changing market requirements. Informa-
tion systems such as manufacturing resource
planning (MRPII) and enterprise resource plan-
ning (ERP) in particular have gained ground in
providing support for achieving an integrated
supply chain.
Firms around the world have been implement-
ing ERP systems since the 1990s to have a uniform
information system in their respective organisa-
tions and to re-engineer their business processes
(Rajagopal, 2002). ERP system as a packaged
software has the advantages of reduced cost, rapid
implementation, and high system quality (Lucas
et al., 1988). Although application packages have
these benefits over custom design software, pack-
aged software have problems of uncertainty in
acquisition and hidden costs in implementation.
Successful ERP implementation must be managed
as a program of wide-ranging organisational
change initiatives rather than as a software
installation effort. Such IT-driven initiatives re-
quire change of the organisation’s socio-economic
system, which is intertwined with technology, task,
people, structure, and culture. Thus organisational
resistance to change is identified as a critical
success factor for ERP implementation (Hong
and Kim, 2002).
Organisational fit and adaptation are important
to implementation of modern large-scale enter-
prise systems that are built with pre-determined
business process methodology. As a result, custo-
misation is a crucial, lengthy, and costly aspect in
the successful implementation of ERP system, and
has, accordingly, become a major speciality of
many vendors and consulting companies. Gefen
(2002) examines how such companies can increase
their clients’ perception of engagement success
through increased client trust, that is brought
about through respective and dependable custo-
misation.
Considering the importance of ERP in SCM, an
attempt has been made in this paper to analyze the
implementation issues of ERP in a major UK
company. The lessons learned from this company
would be useful for other companies in their
efforts to successfully implement modern ERP
system.
2. Enterprise resource planning
In the 1990s innovations in information tech-
nology led to the development of a range of
software applications aimed at integrating the flow
of information throughout a company, and these
commercial software packages were known as
Enterprise Systems. During this period one parti-
cular enterprise system called ERP caught the
attention of some of the worlds largest companies.
It has been estimated that businesses around the
world have been spending almost $10 billion per
year on ERP systems. ERP aims to integrate
business processes through the support of an
integrated computer information system (O’Brien,
1999).
ERP allows the corporate management of a
business, and aims to integrate individual func-
tional systems such as manufacturing, finance,
procurement and distribution. The systems allow
companies to replace their existing information
systems and also help to standardise the flow of
management information and have been regarded
as the next step in the evolution of MRPII. The
MRPII model actually forms the basic core of
ERP and uses similar modules, however some
ERP systems do contain certain modules that were
not originally used within MRPII such as compu-
ter aided design (CAD), distribution resource
planning (DRP), tool management systems
(TMS), and product data management (PDM)
(Yusuf, 1998; Prasad et al., 1999).
ERP uses Internet technologies to integrate the
flow of information from internal business func-
tions as well as information from customers and
suppliers. The system uses a relational database
management system, within client/server network
architecture, to capture valuable management
data. The key principle behind the system involves
entering the data from a series of modular
applications only once. Once stored, the data
automatically triggers the update of all related
information within the system. The systems can
support virtually all areas of an organisation,
across business units, departmental functions
and plants. The development of an ERP system
within a large manufacturing organisation requires
the integration of working practices and the
 ARTICLE IN PRESS
Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
information systems (Davenport, 1998; Mandal
and Gunasekaran, 2002).
Companies that use ERP can gain a competitive
advantage from the way they implement the
system and then exploit the resulting data. Many
companies that have installed ERP have claimed
to be more nimble within the marketplace than
their competitors with hard-to-change custom
made systems (Latamore, 1999).
ERP systems offer companies the following
three major benefits:
* Business process automation.
* Timely access to management information.
* Improvement in the supply chain via the use of
E-communication and E-commerce.
A vital task when implementing an ERP System
is to understand the difference between functions
and modules. Functions are defined as actual
physical tasks that are performed within a
company. Whilst modules can be considered as
pieces of software that help to provide the
functions, different ERP vendors have different
modules that perform the functions. The enor-
mous growth of the Internet and Microsoft
Windows packages, complementary as they are
as collaborative tools has made the argument for
ERP more compelling (Loizos, 1998). It is now a
general industry view that ERP will take them to
new heights of efficiency by enabling them to move
financial and other data speedily from one
department to another (Holt, 1999).
Companies have spent billions of dollars and
used numerous amounts of man-hours installing
elaborate ERP software systems. The ERP soft-
ware vendor market has experienced rapid growth
in the late 1990s. In 1998 there were five major
software vendors offering ERP solutions to
businesses worldwide. The largest of these was
SAP AG (http://www.SAP.com) who earned over
$5 billion in revenue. The Oracle Corp. was the
second largest with $2.4 billion in sales. Followed
in third place by PeopleSoft (http://www.People-
soft.com) who earned $1.3 billion. In fourth place
was J.D. Edwards with $979 million. And finally in
fifth place was the Baan Co. (http://www.Baan.
com) with $743 million in sales (Holland and
Light, 1999).
253
Van Stijn and Wensley (2001) focus on problems
that may arise after ERP systems have been
implemented—the in-use phase. Various problems
have been identified regarding the ERP systems in-
use. Because of the organisational unwillingness or
inability to make technology upgrades (Markus
and Tanis, 2000), the enterprise system may take
on the appearance of a legacy system in disguise.
3. Implementation of ERP
ERP when successfully implemented, links all
areas of a company including order management,
manufacturing, human resources, financial sys-
tems, and distribution with external suppliers and
customers into a tightly integrated system with
shared data and visibility (Chen, 2001). Potential
benefits include drastic declines in inventory,
breakthrough reductions in working capital,
abundant information about customer wants and
needs, along with the ability to view and manage
the extended enterprise of suppliers, alliances and
customers as an integrated whole (Escalle et al.,
1999).
The term ERP stands for enterprise resource
planning, however it is not good enough to just
plan resources required to run the enterprise, they
need to be managed as well. An organisation must
assess itself, to see if it is ready for ERP. It must
determine if it is ready for the competitive business
environment of today and then strengthen its
position for tomorrow’s changes. Some of the
companies that implement ERP systems do not
realise the full benefits that the system offers
because most organisations are not organised in
the correct fashion to achieve the benefits. Many
companies that attempt to implement ERP sys-
tems run into difficulty because the organisation is
not ready for integration and the various depart-
ments within it have their own agendas and
objectives that conflict with each other (Langen-
walter, 2000).
While companies such as Cisco Systems, East-
man Kodak, and Textronix have reaped the
expected benefits of ERP systems, many businesses
are discovering that their ERP implementation
is a nightmare. FoxMeyer Drug, a $5 billion
 ARTICLE IN PRESS
254 Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
pharmaceutical company, recently filed for bank-
ruptcy (Al-Mashari and Zairi, 2000; Chen, 2001).
Dell Computers spent tens of millions of dollars
on an ERP system only to scrap it because the
system was too rigid for their expanding global
operations (Trunick, 1999). ERP implementations
involve, in truth, broad organisational transfor-
mation processes, with significant implications to
the organisation’s management model, organisa-
tion structure, management style and culture, and
particularly, to people (Wood and Caldas, 2001).
ERP software is very adaptable but not very
malleable and companies that wish to use them
correctly have to change their working practices to
fit the software. The key factor of an ERP
implementation is the way in which the software
is configured. The most important issue to identify
before an implementation is the ‘core’ of the
business, which can be identified by the use of the
business model (Chung and Snyder, 2000).
One original motive for buying an ERP system
was to automate business processes, but the
modern view has shifted to the quick access of
up-to-date and timely management information.
The majority of difficulties experienced by ERP
implementations have been the costly development
of additional software to help ‘bridge’ or retrieve
information from legacy systems. Before ERP
management reports can be generated and dis-
tributed to managers, the data has to be created
first which can be a costly and inefficient process.
In an attempt to improve the timeliness and
accuracy of management data, many software
vendors are making end-users responsible for
updating their own information rather than
relying on IT resources.
ERP software consists of a number of modules
that link together to form a complete business
solution, however the main difficulties experienced
by ERP users have been in manipulating the data
stored within the system. Software developers have
begun to address the need for additional informa-
tion tools. Some of these tools include the need for
detailed and advanced planning and advanced
scheduling and customer relationship manage-
ment. Abdinnour-Helm et al. (2003) discussed
the pre-implementation attitudes and organisa-
tional readiness for implementing an ERP system.
Huang and Palvia (2001) analyze the ERP
implementation issues in advanced and developing
countries. Umble et al. (2003) presents the
empirical findings on implementation procedures
and critical success factors for ERP. Van Stijn and
Wensley (2001) address some concerns, methods
and directions for future research on organisa-
tional memory and the completeness of process
modelling in ERP systems.
Most ERP software vendors supply their
customers with an implementation programme as
part of an overall solution package. For example
SAP provides some of its customers with acceler-
ated SAP (ASAP). ASAP suggests the adoption of
a ‘big bang’ implementation. This programme opts
for a quick implementation that is specifically
designed for small and medium sized companies.
‘Big bang’ implementations offer lower costs and
generally use only a few of the software’s inter-
faces, however the risks are greatly increased, as
less time is spent on development and assessing
business needs. There are several more papers on
ERP implementation and some of the most recent
ones include Mabert et al. (2003), Olhager and
Seldin (2003), Umble et al. (2003) and Wood and
Caldas (2001).
4. Systems, applications and products in data
processing
Five former IBM employees originally founded
systems, applications and products in data proces-
sing (SAP) in Mannheim, Germany in 1972. Their
aim was to produce standard software application
programmes that could integrate with each other
to form a business solution. SAP has been
dedicated to produce products that improve the
return on information gathered by an organisa-
tion. The company began its life with the name
‘Systemanalyse Und Programmentwicklung’ and
eventually became known as SAP.
SAP’s first product known as R/2 was built and
prototyped for a subsidiary of ICI. The system
they produced was simply known as system ‘R’,
which stands for ‘Real-time’ processing. This
system was fully integrated and could be used on
the IBM mainframe. The R/2 solution was
 ARTICLE IN PRESS
Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
launched in 1979 and was developed for a
computer mainframe environment, at the time it
was perceived as the most comprehensive system
available to businesses in the world, and it received
great interest from industries in the 1980s.
SAP saw the future potential for the delivery of
information to the end-user via the PC, so SAP
reinvented and developed their product further by
developing a business solution for the client/server
architecture environment, this became known as
R/3 and was released in 1992. In the 1990s SAP
and its R/3 solution would go on to become the
dominant ERP solution, and also become one of
the worlds biggest software houses.
SAP R/3 applications are a range of software
modules. They can be used either alone or
combined to form business solutions. SAP state
that their R/3 applications offer comprehensive
functionality for all standard business needs within
an enterprise. SAP R/3 uses a programming
language called advanced business application
programming (ABAP).
The following are SAP R3’s 12 application
modules: financial accounting, treasury, controlling,
enterprise controlling, investment management, pro-
duction planning, materials management, quality
management, project system, human resource man-
agement, sales and distribution, and plant main-
tenance and service management.
5. Rolls-Royce—case study
In this section, a case study conducted at Rolls-
Royce investigating the implementation of ERP
(SAP) is discussed. The case study starts with
introducing the company and its background,
presenting the status of IT before and after the
implementation of SAP, and giving the detail
chronological phase of the implementation of SAP
in Rolls-Royce. Also, the research methodology is
discussed. The implementation of SAP in the
company and project risks is presented.
5.1. Company background
Rolls-Royce returned to the private sector in
1987 and acquired Northern Engineering Indus-
255
tries in 1989. This acquisition allowed Rolls-Royce
to consolidate its capabilities in the area of
industrial power. A further acquisition was made
in 1995 when Rolls-Royce bought the Allison
Engine Company in the United States, thus
enlarging the company’s presence in aero propul-
sion and industrial gas turbines. The acquisition
allowed Rolls-Royce a major foothold within US
markets. The most modern family of engines is the
Trent series, which are also powerful three-shaft
turbofan engines similar to the RB211, and are
used to power the Airbus A330, A340-500/600.
The Trent series of engines offer greater thrust,
long range flight capabilities and economical
operating and maintenance costs. The Rolls-Royce
Power Generation Market includes both electrical
and nuclear power, which also includes marine
applications, such as providing power plants for
nuclear and naval vessels. Rolls-Royce is a truly
global business offering a range of first class world
leading products. It has facilities over 14 different
countries and also offers first rate after sales
services, covering mechanical overhauls and spare
part distribution. In March 1998 a new organisa-
tional structure was adopted that recognised the
strategy and the need for change to reflect
customer requirements.
5.2. The situations before ERP
Rolls-Royce used over 1500 systems before the
ERP project was started, many of which were
developed internally by Rolls-Royce over the last
two decades. These legacy systems were expensive
to operate and difficult to maintain and develop.
They did not provide accurate, consistent and
accessible data that was required for good and
timely decision-making and performance assess-
ment (e.g. delivery performance, quality metrics).
These ageing systems often did not lend themselves
fully to a modern manufacturing environment.
Some of the legacy systems were so old that they
had year 2000 compliance problems. Work within
Rolls-Royce was functionally orientated and
various departments worked in isolation.
The last major manufacturing system to be
developed and implemented by Rolls-Royce
was MERLIN, which stands for mechanised
 ARTICLE IN PRESS
256 Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
evaluation of resources, logistics and inventory,
the system was basically a scheduling system which
ran on MRPII system principles. The system was
developed in the 1980s and, although it was
capable, it was prone to manual manipulation.
One particular down fall of the system was the lack
of communication between individual sites. MER-
LIN often had difficulty communicating with
another manufacturing system named IBIS, which
stands for inventory based instructing system.
IBIS was an older manufacturing system that
was used at the Bristol and Ansty facilities. Work
in progress was often transferred between sites and
could not be tracked accurately, often causing
inventory and stock take problems.
An additional system named corporate cost
accounting (CCA) was used to financially monitor
transactions, which covered pipeline inventory and
inter-site transport. Rolls-Royce also had a range
of individual systems for controlling and monitor-
ing commercial, financial and procurement func-
tions, these systems had problems interfacing with
each other, as they had different databases and file
formats. The legacy systems did not allow Rolls-
Royce to establish direct, on-line communication
with customers, partners and suppliers. In fact,
these systems did not support significant growth of
the business and were not sufficiently agile to keep
pace with the changing business environment.
5.3. IT at Rolls-Royce
In 1996 Rolls-Royce formed a partnership with
electronic data services (EDS). The Rolls-Royce
IT department was outsourced to EDS, which
meant that EDS were responsible for the develop-
ment of the company’s IT systems as well as taking
over the existing structure and providing adequate
IT resources. This move was made in order to
allow Rolls-Royce to concentrate its efforts on its
main area of expertise—the making and selling of
aero engines. Rolls-Royce decided that a partner-
ship with a world leading IT outsourcer would
benefit the company far more than designing and
maintaining their own IT systems. EDS were
chosen because of their substantial experience
within the aerospace industry. EDS also had the
responsibility for employing specialist consultants.
In 1998 changes were made to flatten the
structural hierarchy. The customer focused busi-
ness units (CFBU) were made responsible for
making sales deals within the various market
segments. Whilst the operational business units
(OBU) formed the manufacturing support for
producing the product, the Executive Group
controls the whole business and makes decisions
on the overall direction of the company.
5.4. The implementation project
The ERP project consists of a management team
of specialists from the external outsourcing com-
pany EDS. EDS also have the specialised talents of
SAP consultants. Within the project team are
specialist internal managers and staff that have
vital knowledge of cross-functional business rela-
tionships and experience of the old internal
systems. In conjunction with this team each OBU
has its own ERP planning team, which is
responsible for implementing working changes
and training. The project implementation pro-
blems can be grouped into three areas of cultural,
business and technical difficulty.
5.4.1. Cultural problems
The implementation project team expected a
high acceptance of the system in areas that provide
just as good or better functionality than the old
system. However some functions and processes
might not get the full appreciation the legacy
systems once had. The project team decided to
resolve this by illustrating the improvements made
to the company as a whole, thus breaking the
traditional segregation of OBUs and departments.
The original implementation plan was increased in
an attempt to address training and cultural
changes. Training took the form of organised
seminars, which were split into two distinct groups
of specialists and mass users. The specialist
training was carried out and conducted by SAP
and was technically based. These specialist experts
then in turn trained expert users. The remaining
training for end-users was conducted internally in
collaboration with EDS consultants. The training
carried out within the seminars was supported by
demonstrations within the workplace, along with
 ARTICLE IN PRESS
Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
information meetings and presentations to relay
information to all employees about the changes of
working practices. In all, more than 10,000 people
would have been trained.
5.4.2. Business problems
SAP R/3 requires a fairly rigid business struc-
ture for it in order to work successfully. The
participants of cross-functional workshops soon
understood that their working practices must be
adjusted in order to fit SAP, ultimately changing
the way Rolls-Royce does business. They achieved
this by using an internal business process re-
engineering (BPR) programme. The programme
consisted of four steps, the first involved drawing
and mapping the current processes. The second
step involved identifying any problems or issues
raised from the mapped process. The third step
involved applying some of these issues to a
demonstration of SAP, to identify potential
problems within the new system. The fourth step
involved the re-mapping or modification of the
processes in line with SAP. The modifications to
the Rolls-Royce business process meant that the
SAP R/3 software need not be modified. Mod-
ifications to the software would have been
extremely expensive both in terms of implementa-
tion resources and the fact that newer software
versions would be difficult to install in a modified
Build Create
Customer Customer
Relationship Solutions
Plan the Resource the
Business Business
Satisfy the Manage
Shareholder Cash
Fig. 1. Business process model.
257
system. SAP named this unmodified software
implementation ‘Vanilla SAP’.
5.4.3. Technical problems
The main technical problems that Rolls-Royce
has encountered have been with the accuracy of
data. The new system requires the retrieval of old
data from the legacy systems that has to be
normalised, screened and stored in a sensible data
format within the new systems data repository.
The duplication of data was a major concern that
Rolls-Royce had to address. In some special areas
the old systems was kept running until such time as
they could be phased out by the new systems, and
to do this EDS built interfaces between the
systems. The CAD system used by Rolls-Royce
remained the same, as the process to alter the file
formats would be too expensive and require use of
valuable resources that are needed for the core
implementation.
Rolls-Royce has nine principal business pro-
cesses, which when taken together describe every-
thing the company does. Fig. 1 is a schematic
representation of the business processes and the
interfaces.
Rolls-Royce decided to adopt and utilise the
SAP solution offered for the aerospace and
defence industry. The SAP aerospace and defence
industry solution is the market leader in its
Resolve
Customer
Problems
Generate Fulfil
Orders Orders
 ARTICLE IN PRESS

258 Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266

industry and is highly configurable for flexible
‘vanilla’ implementation. Predetermined imple-
mentation points from the Rolls-Royce Steering
Committee and Implementation Team defined the
release strategy for the project. Any future third
party software products must first be accredited by
SAP to safeguard the upgrade process and would
require a justified business case. Business reports
that are generated by SAP have to be fully justified
in a business case, which follows a standard format
for internal use. Data entering the project has to be
identified, validated, cleaned, loaded, archived and
then maintained within a Data Warehouse.
Rolls-Royce have estimated that over 1000
additional PCs will be required and the total cost
for the network infrastructure was about two
million pounds. The company required over 6000
SAP licences for users across all the business. The
server was provided by Sun Microsystems and in
excess of 2 Terabytes of disk space. The system
required almost 35 weekly MRP runs cascaded by
plant. A UNIX server bridges the data from legacy
systems and testing and training required an NT
server. The detail implementation model plan with
project time-scale is shown in Fig. 2.
5.4.4. Phase 1 (strategy and direction)
The first phase of the project was a short
intensive study to set the scope of the project
and provide an outline plan and costing. A
steering committee was formed to administer the
financial guidance of the project and a ‘ERP Core
Team’ was formed to control and oversee the
actual implementation process.
5.4.5. Phase 2
During the second phase a detailed plan was
created and a prototype system was installed. An

1998 1999 2000 2001

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1

Phase 1
1

2 SAP
Phase ‘Go Live’
2
3

4 5
Phase
3
6

Key

Phase 1 1 Strategy & Direction

2 Planning Analysis & Convergence

Phase 2
3 Early Deployment

4 Wave 1 – Focus on Operations

Phase 3 5 Wave 1 – Pilot

6 Wave 2 – Focus on Assembly & Spares

Fig. 2. ERP implementation model with project time-scale.

 ARTICLE IN PRESS
Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
enterprise model was developed based on the
Rolls-Royce Allison model, and all the existing
projects within Rolls-Royce were drawn together
and merged. Many issues were addressed including
the integration of Better Performance Faster
(BPF) initiatives. A series of workshops known
as ‘High Level Process Confirmation Workshops’
took place, which involved over 200 line personnel
and the ERP core team, and took into considera-
tion the various business processes within the
scope of the project. These workshops were closely
followed by ‘Business Simulation Workshops’,
which involved approximately 300 line personnel
and were used to forge a strong relationship
between the ERP core team and line personnel
and avoid possible pitfalls, such as initiative
fatigue or lack of co-operation. Activities carried
out during the second phase of the project
included:
* Preliminary design review—developing a design
and implementation strategy, defining the scope
of the project, and developing the business
process model.
* High level design review—analyse the enterprise
model, and develop ‘Vanilla’ prototype.
* Critical design review—detailed design and
customisation of the prototype.
* Implementation realisation—integration test-
ing.
* Technical/operation review—user acceptance
testing.
* Post implementation review—system deploy-
ment, systems conversion, user training before
the ‘Go Live’.
During phase two the projects core structures
were identified. Integrated programme manage-
ment (IPM) was also adopted for research and
development and would eventually cover the
whole business. Additional activities included the
support of finance and staff work booking. Phase
two was completed at a cost of d5.2 million, within
two weeks of the plan.
5.4.6. Project changes
During phase two a significant change was made
from the original timing of phase one, the
259
completion of wave one was deferred for about 6
months. This has resulted in a knock-on delay to
wave two by a corresponding amount. The change
in schedule was possible without a significant
increase in cost because the problems were
addressed early enough in the programme. There
were four main reasons for the change:
* To give the line organisations more time to
prepare, train and clean up data.
* To provide an additional 5 months period for
pilot running and early development of the
system.
* To provide additional time for the completion
of other pre-requisite projects being managed
by BPF. Specifically the deployment of product
data manager (PDM) and shop floor data
manager (SFDM) on which SAP is critically
dependent.
* To provide additional time for resolving diffi-
culties with successful use of SAP at RR
Allison.
5.4.7. Phase three (implementation)
This phase was too large to implement in one
go, and thus was divided into two ‘waves’. Both
waves were concerned with the physical imple-
mentation of the system and its architecture. The
waves were also concerned with changing working
practices within the company.
5.4.8. Wave one
This wave was concerned with the replacement
of legacy systems. IPM was also introduced for
new production projects during wave one. The
new manufacturing execution system, known as
shop floor data management (SFDM) was also
introduced during wave one. The ultimate end to
wave one was a SAP pilot project at one of Rolls-
Royce facility. The pilot laid the foundation for
the full ‘go live’ throughout the company about a
year afterwards. The first wave had the ultimate
aim of providing new capabilities for gas turbine
operations.
5.4.9. Wave two
The second wave was approximately 1 year in
duration, and was not operational until after the
 ARTICLE IN PRESS
260 Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
first wave finishes. The second wave was concerned
with implementing engine assembly, spares, logis-
tics and human resource elements within the
project. By this time the legacy systems was
switched to ‘view only’ as SAP becomes the
executive system. Once the new system shows a
positive response the older systems was phased
out. IPM completely covered the whole business
by the end of the second wave.
5.5. Changes to the existing system
During phase two, modifications were made to
the legacy systems. These modifications were
adopted in a series of suites.
5.5.1. Suite 1
Plan the supply chain: This takes place as part of
the corporate business planning activity with a 2–5
year horizon. A review was undertaken of all
potential future sales, including engines and
spares, identifying the probable minimum and
maximum levels. The supply chain capacity
was compared to the range of possible sales
scenarios.
Master schedule key programmes: Suite 1 sup-
ports the sales and operating review board
(SORB), which is a director level meeting, which
strategically plans engine sale and factory capacity
on a 2–5 year rolling forecast. The SORB makes
decisions affecting changes to achieve capacity in
terms of manpower, machines, technology, and
ultimately for factories to meet the potential sales
forecast. The SORB process has often been
referred to as the ‘Evaluation and Commitment
Acceptance Program’. The SORB meet every 3
months to decide what has to be made and when.
The SORB records baseline data from the previous
meeting and then identifies new changes. For
example if 20 engines were required last, it does
not necessarily mean that 20 were actually built, a
particular manufacturing unit may have had
machine breakdowns, which could have effected
the amount of work leaving the factory. A copy of
the plan gets taken from the Project System
module within SAP and then the new requirements
are input into an inactive version of the SAP
module Demands Management. This copy can
then be used in ‘what if’ simulations before finally
being transferred to the master copy as the latest
SORB.
5.5.2. Suite 2
Plan and schedule the factory: This converts the
agreed schedule from suite 1 into a production
plan and enables all the manufacturing units to
plan capacity to produce the required components.
The production plan may include schedule
smoothing. Schedule smoothing is a process,
which converts erratic customer requirements into
a consistent production plan that allows the
facility to operate efficiently on a regular pattern.
The operations businesses will have the responsi-
bility of holding excess inventory that is created by
schedule smoothing.
Schedule the shop: This converts the production
plan into a detailed shop plan. It generates when to
launch material (either raw material or part-
finished details) onto the shop floor, and when
the material should meet identified stages of the
manufacturing process.
5.5.3. Suite 3
Operate the factory: Suite 3 covers the control of
workflow through the shop from the initial
generation of launch paperwork, right through to
delivery of finished products. It gathers informa-
tion on the booking of work at identified stage
points in the manufacturing process, and collects
operating data for cell level management. It
controls inventory between the manufacturing
units and controls the flow of components through
the make process. Once SAP has become executive
these suites will be replaced by SAP modules and
SFDM, which will handle and perform these
tasks.
The complete systems architecture for the
project is extremely complicated, however for the
purpose of this discussion a brief overview can be
seen in Fig. 3. The core business operations are
supported by SAP and integrate with other
strategic software products (highlighted in brack-
ets). Rolls-Royce are using 11 out of 12 SAP
modules. They decided not to use the plant
maintenance module as they already have an
adequate system called MAXIMO (Fig. 4).
 ARTICLE IN PRESS

Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266 261

Plan the supply chain

Suite 1 APS – Enterprise Level

Master Schedule Key

Programmes

Plan & Schedule the

Factory
APS – Factory Level and
Suite 2
SAP

Schedule the Shop

Suite 3 Operate the factory SFDM and SAP

Fig. 3. The five box diagram of suite implementation.

5.6. The ERP pilot Saturdays and Sundays as well!’’—Core Im-

A small-scale pilot of the system was run for 3
months and throughout this period, a facility
known as number 4 shop, which was part of the
transmissions and structures operations unit be-
came the central focus of attention for the whole
company. This facility was chosen for the pilot run
because the facility only produced 280 parts, and
material flowed into the facility at low volumes
from external suppliers and internal operational
units. The purpose of the pilot was to demonstrate:
* business principles;
* processes;
* procedures;
* role definitions and behaviours;
* software, hardware and data transfers.
‘‘We initially looked at over 1000 part numbers,
identified which ones had a schedule against
them, its the ones that have a schedule that we
did the data clean up on, which came down to
280 part numbers. If we had to do it across the
1000 part numbers we would never have got it
finished. For each part number there are
around about 30/40 operations, there had been
7 of us working full time on it, that’s included
plementation Team Member (a).
The implementation of ERP has created two
new roles at Rolls-Royce, these key roles were:
* MRP controllers;
* Capacity owners.
‘‘We had to go through symmetric tests,
aptitude tests, interviews, it was quite daunting
really, I mean to think that here I am, quite
settled but I’ve got to make these moves. I’m
told that I’ve got to spend X amount of time
around a PC, it’s a bit strange as I must spend
around 80 percent of my time on the shop floor
to day, and that’s going to change dramatically.
I will own the men and machines, the capacity
y’’—Core Implementation Team Member (b).
‘‘An MRP Controller is going to be in charge of
the inventory, and getting the raw material in,
and talking to the supplier and the customer.
But they’ve also got to release the material to
the shop floor, and then release the material
when its finished as a component to the
customer.’’—Core Implementation Team Mem-
ber (c).
 ARTICLE IN PRESS

262 Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266

Finance Programme Management Human Resources

Financial Core Organisation & Finance Research & Development Resourcing
Consolidation Structure Series Production Compensation & Benefit
(Hyperion) Asset Management Spares & Other Payroll
Accounts Payable Profit & Cost Accounting Health & Safety
Accounts Receivable Staff Work Booking Organisation Development
Bill of
Product & Inventory Accounting Employee Development
Materials
Profitability Analysis Pensions

Product & Process ERP (SAP R/3) Supply Chain Planning

Product Data Development Sales & Operations Planning
Management Assembly Configuration Demands Management
Material Integrated Business Processes &
Master Basic Component Configuration Master Production Scheduling
(Metaphase) Systems Applications
Data Assembly Process Planning Material Requirements Planning
Component Process Planning Forecasting & Provisioning
Supply Chain Optimisation

Procurement & Inventory Manufacturing & Assembly Order Management

Sourcing Execution Sales Order Administration
Procurement Manufacturing Execution
Inventory Management Assembly Operations Service Management
(Manufacturing) Tool Management
Product As-built Inventory Management Plant Management
Structure Data (Assembly)
Warehouse Management

Kits & Tools

Requirements Routings Work Work Status
Launch

Assembly Instruction & Control (CAPP) Manufacturing

Process Planning Execution (SFDM)
Execution

Facilities & Services

Management (MAXIMO)

Fig. 4. Systems applications architecture overview.

MRP Controllers and Capacity Owners were
sent on a residential behavioural course for 2
weeks and for most of them this was their first
glimpse at how they would be working in the
future.
‘‘We’ve got 20 users at Ansty, in the pilot and to
connect those users we’ve had to install a lot of
fibre network throughout the Ansty site. Its
such a huge task of delivering the infrastruc-
ture, that’s the reason why people have been so
frustrated, saying things like, I go live in 4
weeks and I haven’t got a piece of kit yet!
Once they’ve been trained and know how to
use that piece of kit they will have it there
and then. The second thing to bear in mind
is to really understand what roles the user is
going to play in the pilot. This is to ensure
that we have the right access permission
correctly set-up for the user so they can use
SAP correctly.’’—Core Implementation Team
Member (d).
Systems testing and getting the user to accept
the system were important roles undertaken in the
 ARTICLE IN PRESS
Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
pilot. The following quotes give some insight:
‘‘Before the system can go-live the team needed
to check that it works properly in their
operational environment. User acceptance
training is all about buy off, we’re trying to
make sure that the ownership is with the
business not the core team. They have to say
we’re happy!’’—Core Implementation Team
Member (e).
‘‘I thought that when we first started the
programme it would be pretty black and white,
finishing the unit testing, starting the integra-
tion testing, starting the user acceptance testing,
its not that simple! What you have are different
bits of either configuration or bridges that you
have to phase in at different parts of the testing
cycle. They cannot be missed but they cannot be
finished in time to say, I’ve finished all unit
testing, I’ve finished all integration testing its
actually quite blurred.’’—Core Implementation
Team Member (e).
Loading clean data into the new system also
produced many difficulties. The following quotes
give some insight:
‘‘Its not until you actually try and load the data,
get the data in the system and then hopefully let
it feed into SAP that you really start to
understand that things aren’t quite right. We’ve
lost something like a week and a half so far
trying to get the bill of materials into metaphase
and then feed them into SAP.’’—Core Imple-
mentation Team Member (e).
‘‘We are going to load into SAP about 2 percent
of the part numbers that we would actually load
in wave one, and we are finding it incredibly
difficult!’’—Core Implementation Team Mem-
ber (d).
The initial problems experienced on ‘going live’
were:
* User authorisation problems, such as, password
and user level clearance.
* Work was temporarily halted on the shop floor,
as route cards were unavailable.
263
* Values between the systems were incorrect, so
comparisons were made on the values from the
legacy system with those on the new system,
such as inventory levels and WIP.
* Transaction problems occurred from the first
MRP run, so comparisons were made, between
the old and new systems, and corrections were
made.
A second pilot was also carried out for non-
production purchasing. The second pilot ran
executively, covering Derby-based purchasing of
ground support equipment. A third pilot also
was run by the Airline Business. The third pilot
was non-executive, but designed to specifically
explore the interplay between Metaphase PDM
and SAP.
5.7. The ‘Go Live
As the main ‘Go Live’ of the new system was
planned, the most difficult part of the cut over
process was in transferring the data from legacy
systems. The shear volume of data that has to be
transferred is far greater than any normal transac-
tion load that will be carried out by the system
thereafter. In order for this process to be successful
the data must be kept in a ‘stable’ state for a
period of roughly 10 weeks. The initial data to be
transferred includes some transaction data and
master data, for example, lists of suppliers. If any
changes occur to the data on the old systems after
the transfer, they are logged and then passed
through to the new system. The remaining data
was loaded in after the ‘Go Live’.
The next step during the ‘Go Live’ process
involved running the MRP system to initialise the
whole system. Purchase orders and purchase
requisitions was not transferred from the old
system, instead the MRP run should create them
fresh. The whole ‘Go Live’ process took roughly 2
weeks to complete, and during this time the new
system was ‘off the air’.
Immediately after the ‘Go Live’ the existing
legacy systems was switched to view only mode.
The view only mode enabled comparisons to be
performed between the old and new systems.
However, the legacy systems ceased to be executive.
 ARTICLE IN PRESS
264 Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
5.8. Project risks
The ERP Project at Rolls-Royce covers many
different departments and many different topic
areas, all of which have associated risks. In order
to address and take positive action to avoid failure
or potential errors the ERP implementation team
maintained and recorded in a great detail, a risk
register. Every issue within the company, which
involves risk has been catalogued and continu-
ously reviewed. The risk register is very large,
however the Rolls-Royce ERP Intranet page offers
a brief summary of some of the major risks:
* The possible failure or inability to align goals
through conflicting directions within the orga-
nisation.
* The non-delivery or non-availability of reliable
IT hardware and infrastructure both before and
during implementation.
* The possible failure of providing inadequate
and ongoing support after implementation,
from both Rolls-Royce and EDS.
* The resistance of change to new process
methods by management and supervision.
* Management and supervision may treat the
project as merely an IT implementation, rather
than change in process methods.
* Inadequately educating the workforce to oper-
ate the new system properly.
* Possible failure to cut over to the new system
through an inability to load data.
* Possible failure to cut over to the new system
through the inappropriate systems testing of
volume, stress and data conversion.
* Possible failure to give ERP adequate priority
due to the number of existing and ongoing
business improvements.
* Maintenance difficulties may occur on bridged
legacy systems.
* The project may impact on company interim
and end of year accounts.
* The PDM project may not be sufficiently
positioned in time with the ERP project.
* Possible changes to kitting demand during ‘go
live’ may stretch the new system and those
operating it on a learning curve beyond
capacity.
* The decision to implement Wave 1 separately
from Suite 3 may fail to integrate the new
systems.
* Airline Business After-sales may not be able to
analyse and manipulate inventory investment in
stock target groups (MERLIN functionality
which helps to control forecasting for
Airline Spares stock targets will be removed in
Wave 1).
6. Summary and conclusions
Rolls-Royce has a large complex business
process and the project has had to assess the
effects throughout the whole business, which is
equivalent to ten medium sized companies pulling
together as one. This has caused administrative
difficulties, particularly in the first phase of the
project, whilst setting the strategy and overall
direction. Rolls-Royce decided to make these
radical changes to their business, in response to
increased orders from the market place, and also
from the fact that ERP has become a standard
solution world-wide within the Aerospace and
Defence industry. The introduction of SAP R/3 at
the facility in the USA was a major factor in
influencing the UK implementation. Rolls-Royce
produce a range of quality world class turbine
engines, and have recognised that they must
change in order to compete effectively with their
competitors. Accurate information systems and
direct communication with suppliers are vital
when offering customers a committed promise to
deliver.
Rolls-Royce has understood the business, cul-
tural and technical difficulties of such a large
project, and has developed a solid core implemen-
tation team. The team has used the specialist skills
of consultancy specialists. The partnership with
EDS has produced a sound architectural frame-
work for the project, thus allowing Rolls-Royce to
concentrate its efforts on manufacturing turbine
engines. A project of this size would never run
smoothly and difficulties have occurred through-
out the implementation and will no doubt occur in
the future. The company have taken a different
approach to IT systems but have not let the project
 ARTICLE IN PRESS

Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266 265

become just another IT system. The core imple- data is stored centrally and is extracted from
mentation teams have taken into account the operational, historical and external databases. The
needs of both the managerial and end-user. The data is first screened then edited and finally
following list contains just some of the problems standardised for future retrieval. The data is
encountered: stored in a logical user-friendly format. It allows
non-technical users to create database queries
* Matching the process to the software config-
allowing the simple retrieval of management
uration.
information for business intelligence and manage-
* Training people to accept change, and getting
rial decision making. The database continually
them to do business in a totally new way.
absorbs new data and integrates it with the
* Teaching employees to use modern IT equip-
previous data.
ment.
The full benefits of the project will not be fully
* Equipment not delivered on time, or delays in
experienced or achieved until the system becomes
technical equipment installation.
executive and has a period of stability, for at least
* Data clean up has been particularly time
a whole year. Once the system has become stable
consuming as many legacy systems have been
and users have had time to adjust to new working
involved.
practices the benefits of lower IT cost will become
* Training the behaviour of SAP users such as
visible. An immediate benefit that will be achieved
MRP Controllers and Capacity Owners.
by the system will be the ability to promise and
then deliver to the customer on time. This was
Many activities have taken place, which have something that the older systems could never
been vital to the overall success of the project, such achieve, as they often used due dates that were in
as: the past. SAP can only use current information.
The ability to deliver on time will improve
* Bridging the legacy systems and cleaning up
customer satisfaction and also improve customer
suspect data has given the company more trust
confidence, which should lead to an increase of
in its management of information.
orders in the future. The system will also improve
* Training senior management, particularly the
the relationship in the supply chain, as transac-
executive group, who are responsible for the
tions will be made easier via the use of Electronic
overall direction of the company and are not
Communications.
technically orientated.
The sustainability of enterprise information
* Managing effective relationships and leading
systems (EIS) during the post-implementation
teams in both technical and non-computer
period needs to be looked into. There is a lack of
based environments.
clear understanding about the strategic needs and
* Manufacture simulation exercises.
requirements for sustaining the effectiveness of
* Transactional training.
large-scale information systems after a period of
* Shop floor communication with line workers
relative stability following initial implementation.
was an exercise that occurred during the
Sustainability management of EIS is therefore a
implementation of suite 3. This required line
very important research dimension that needs to
workers to attend workshops to learn new PC
be explored to maximise the benefits of an
skills in order to book work.
expensive information system investment such as
ERP.
SAP guarantee that newer versions of their
software will upgrade SAP reports, whilst specially
created reports will require re-writing of the Acknowledgements
software. The future of the project will eventually
lead to the need for a Data Warehouse. A Data The authors are very grateful to two anony-
Warehouse is an integrated collection of data. The mous referees for their constructive and helpful
 ARTICLE IN PRESS
266 Y. Yusuf et al. / Int. J. Production Economics 87 (2004) 251–266
comments which helped to improve the presenta-
tion of the paper considerably.
References
Abdinnour-Helm, S., Lengnick-Hall, M.L., Lengnick-Hall,
C.A., 2003. Pre-implementation attitudes and organiza-
tional readiness for implementing an enterprise resource
planning system. European Journal of Operational Re-
search 146 (2), 258–273.
Al-Mashari, M., Zairi, M., 2000. The effective application of
SAP R/3: A proposed model of best practice. Logistics
Information Management 13 (3), 156–166.
Chen, I.J., 2001. Planning for ERP systems: Analysis and
future trend. Business Process Management Journal 7 (5),
374–386.
Chung, S.H., Snyder, C.A., 2000. ERP adoption: A technolo-
gical evolution approach. International Journal of Agile
Management Systems 2 (1), 24–32.
Davenport, T., 1998. Putting the enterprise into the enterprise
system. Harvard Business Review 76 (4), 121–131.
Escalle, C.X., Cotteleer, M.J., Austin, R.D., 1999. Enterprise
Resource Planning (ERP): Technology Note. Harvard
Business School Publishing, Boston, MA.
Gefen, D., 2002. Nurturing clients’ trust to encourage engage-
ment success during the customization of ERP systems.
Omega 30, 287–299.
Holland, C., Light, B., 1999. A critical success factors model for
ERP implementation. IEEE Software (May/June) 30–35.
Holt, L., 1999. Competition heats up in ERP. Info World 21
(6), 65.
Hong, K.-K., Kim, Y.-G., 2002. The critical success factors for
ERP implementation: an organizational fit perspective.
Information & Management 40, 25–40.
Huang, Z., Palvia, P., 2001. ERP implementation issues in
advanced and developing countries. Business Process
Management Journal 7 (2), 276–284.
Langenwalter, G.A., 2000. Enterprise Resource Planning and
Beyond—Integrating Your Entire Organization. St. Lucie
Press, Boca Raton, FL.
Latamore, G., 1999. Flexibility fuels the ERP evolution.
APICS—The Performance Advantage, October 44–50.
Loizos, C., 1998. ERP: is it the ultimate software solution?
Industry Week 7, 33.
Lucas, H.C., Walton, E.J., Ginzberg, M.J., 1988. Implementing
packaged software. MIS Quarterly 537–549.
Mabert, V.A., Soni, A., Venkataraman, M.A., 2003. Enterprise
resource planning: Managing the implementation process.
European Journal of Operational Research 146 (2),
302–314.
Mandal, P., Gunasekaran, A., 2002. Application of SAP R/3 in
on-line inventory control. International Journal of Produc-
tion Economics 72, 47–55.
Markus, M.I., Tanis, C., 2000. The enterprise systems
experience—from adoption to success. In: Zmud, R.W.
(Ed.), Framing the Domains of IT Research: Glimpsing the
Future Through the Past. Pinnaflex Educational Resources,
Inc., Cincinnati, OH, pp. 173–207.
O’Brien, J., 1999. Management Information Systems. McGraw-
Hill Inc., London.
Olhager, J., Seldin, E., 2003. Enterprise resource planning:
Survey of Swedish manufacturing firms. European Journal
of Operational Research 146 (2), 365–373.
Prasad, B., Sharma, M.K., Godla, J., 1999. Critical issues
affecting an ERP implementation. Information Systems
Management 16 (3), 7–14.
Rajagopal, P., 2002. An innovation-diffusion view of imple-
mentation of enterprise resource planning (ERP) systems
and development of a research model. Information &
Management 40, 87–114.
Trunick, P.A., 1999. ERP: Promise or pipe dream? Transporta-
tion & Distribution 40 (1), 23–26.
Umble, E.J., Haft, R.R., Umble, M.M., 2003. Enterprise
resource planning: Implementation procedures and critical
success factors. European Journal of Operational Research
146 (2), 241–257.
Van Stijn, E., Wensley, A., 2001. Organizational memory and
the completeness of process modelling in ERP systems:
Some concerns, methods and directions for future research.
Business Process Management Journal 7 (3), 181–194.
Wood, T., Caldas, M.P., 2001. Reductionism and complex
thinking during ERP implementations. Business Process
Management Journal 7 (5), 387–393.
Yusuf, Y., 1998. An empirical investigation of enterprise-wide
integration of MRPII. International Journal of Operations
& Productions Management 18 (1), 66–86.